Skip to content

feat: put image providers on one contract and drop CLI provider branches #355

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
kiyeonjeon21 merged 1 commit into from
Oct 5, 2026
Merged

feat: put image providers on one contract and drop CLI provider branches #355

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 6 additions & 4 deletions AGENTS.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -129,10 +129,12 @@ General expectation:
After editing the catalog, run `pnpm gen:models` (MODELS.md is partly
generated).
- Video providers implement `VideoGenerator`
(`packages/ai-providers/src/video/`): HTTP goes through `providerRequest`,
failures are `ProviderError`s with a `kind`, and each provider passes
`describeVideoContract`. Callers branch on `error.kind`, never on provider
wording.
(`packages/ai-providers/src/video/`) and image providers implement
`ImageGenerator` (`src/image/`): HTTP goes through `providerRequest`,
failures are `ProviderError`s with a `kind`, and each provider passes its
contract suite. The CLI opens providers through `createVideoGenerator` /
`createImageGenerator` and branches on `error.kind`, never on provider
names or wording.

## Conventions

Expand Down
22 changes: 11 additions & 11 deletions docs/cli-reference.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -491,7 +491,7 @@ Cost tier: _not tagged_

#### `vibe generate image`

Generate image using AI (Gemini, OpenAI gpt-image, Grok, or Runway)
Generate image using AI (OpenAI GPT Image, Gemini, or Grok)

Product surface: `public`

Expand All @@ -500,15 +500,15 @@ Cost tier: `high`
**Parameters:**

- `prompt` _(string)_ - Image description prompt (interactive if omitted)
- `provider` _(string)_ _(openai \| gemini \| grok \| runway)_ - Provider: openai (default when OPENAI_API_KEY set), gemini, grok, runway
- `apiKey` _(string)_ - API key (or set env: OPENAI_API_KEY, GOOGLE_API_KEY)
- `output` _(string)_ - Output file path (downloads image)
- `size` _(string)_ _(default: `"1024x1024"`)_ - Image size (openai: 1024x1024, 1536x1024, 1024x1536)
- `ratio` _(string)_ _(default: `"1:1"`)_ - Aspect ratio (gemini: 1:1, 1:4, 1:8, 4:1, 8:1, 16:9, 9:16, 3:4, 4:3, etc.)
- `quality` _(string)_ _(standard \| hd)_ _(default: `"standard"`)_ - Quality: standard, hd (openai only)
- `style` _(string)_ _(vivid \| natural)_ _(default: `"vivid"`)_ - Style: vivid, natural (openai only)
- `provider` _(string)_ _(openai \| gemini \| grok)_ - Provider: openai (default when OPENAI_API_KEY set), gemini, grok
- `apiKey` _(string)_ - API key (or set env: OPENAI_API_KEY, GOOGLE_API_KEY, XAI_API_KEY)
- `output` _(string)_ - Output file path
- `ratio` _(string)_ _(default: `"1:1"`)_ - Aspect ratio: 1:1, 16:9, 9:16, 4:3, 3:4, ... (OpenAI maps it to its nearest size)
- `size` _(string)_ - Explicit OpenAI size: 1024x1024, 1536x1024, 1024x1536 (overrides --ratio)
- `quality` _(string)_ _(low \| medium \| high)_ - Quality: low, medium, high (OpenAI and Grok; standard/hd still accepted)
- `resolution` _(string)_ - Gemini 1K/2K/4K (2K and 4K need Pro) or Grok 1k/2k
- `count` _(number)_ _(default: `1`)_ - Number of images to generate
- `model` _(string)_ - Model. Gemini: flash (default), lite, pro. OpenAI: 2.5 (default), flare, 2, 1.5. Grok: pro
- `model` _(string)_ - Model. OpenAI: 2.5 (default), flare, 2. Gemini: flash (default), lite, pro. Grok: pro
- `dryRun` _(boolean)_ - Preview parameters without executing

#### `vibe generate motion`
Expand Down Expand Up @@ -777,7 +777,7 @@ Cost tier: `high`
- `provider` _(string)_ _(gemini \| openai \| grok)_ _(default: `"gemini"`)_ - Provider: gemini (default), openai, grok
- `apiKey` _(string)_ - API key (or set env variable)
- `output` _(string)_ _(default: `"edited.png"`)_ - Output file path
- `model` _(string)_ _(default: `"flash"`)_ - Model: flash, lite, pro (Gemini); pro (Grok)
- `model` _(string)_ - Model: flash (default), lite, pro (Gemini); 2.5, flare, 2 (OpenAI); pro (Grok)
- `ratio` _(string)_ - Output aspect ratio
- `size` _(string)_ - Resolution: 1K, 2K, 4K (Gemini Pro only)
- `dryRun` _(boolean)_ - Preview parameters without executing
Expand Down Expand Up @@ -1256,7 +1256,7 @@ Cost tier: `free`
- `insertInto` _(string)_ _(default: `"index.html"`)_ - Root composition file to update
- `project` _(string)_ _(default: `"."`)_ - Project directory
- `noStoryboard` _(boolean)_ - Do not sync STORYBOARD.md; insert this scene directly into the root composition only
- `imageProvider` _(string)_ _(gemini \| openai)_ _(default: `"gemini"`)_ - Image provider: gemini, openai
- `imageProvider` _(string)_ _(gemini \| openai \| grok)_ _(default: `"gemini"`)_ - Image provider: gemini, openai, grok
- `tts` _(string)_ _(auto \| elevenlabs \| openai \| kokoro)_ _(default: `"auto"`)_ - TTS provider: auto, elevenlabs, openai, kokoro (default auto - ElevenLabs key > OpenAI key > Kokoro local)
- `voice` _(string)_ - Voice id (ElevenLabs name/id, OpenAI voice like marin, or Kokoro id like af_heart)
- `noAudio` _(boolean)_ - Skip TTS even when --narration is provided (useful for tests/agent dry runs)
Expand Down
Loading
Loading